Crankshaft bearings thermal and oil flow...

[Shannow]

http://www.ukintpress-conferences.com/uploads/SPKEX13/Day1_4_Omar_Milan.pdf

Mahle conference paper on thermal and oil flows in crankshaft bearings, plus changes in design to reduce lubricant pump capacity.

Some pertinent points:

Page 3 shows a "safe operation range"...(for explaination, that line to the right "oil film stability" is where the film is typically too thick for the load, and things like whip and whirl take place)

Page 5 gives indicative pressures in a bearing...note that a 60psi oil pump can't compete with the pressures in the oil wedge...the oil pump is only supplying oil to the bearings, the bearings draw from that supply what they need.

Page 6. heat generated through shearing in the mains, flows into the block (plus carried away with side leakage and convection to windage).

Page 6. heat generated through shearing in the big ends flows through the crank and into the mains (plus carried away with side leakage and convection to windage).

Page 9. Shows measured oil flows and temperature rises across bearings, both plain, semi grooved, and full grooved. Flow and temperature rise with RPM.
* Full grooved bearings flow more, and have less temperature rise. (Reason being, and it's not explained in the paper is that a full grooved bearing sits "lower", as it're really two very narrow bearings...holds less load than a plain shell...and the clearance at the top (feed end) is greater, letting more oil out.
* Do some mental arithmetic, and although the full grooves flow more, and run cooler, the delta T X Volume means that about the same overall heat needs to be dissipated with either strategy...one uses a LOT less oil in circulation 'though.

P10/11, shows crank drilling changes on a 3 cylinder engine that allows changes in bearing design, offering a 15% reduction in crank oil supply requirement.

P12 to 14, shows oil distribution volumes in a 4 cylinder diesel. Then same crank drilling changes as previous section, and measured results...45% reduction in crank flows possible through drilling/bearing changes.

Page 15 some flow to big ends. A little ambiguous, but I think the left chart "measured" is a centre drilled oil supply, and the red "measured" on the right with a traditional oiling, showing drop off at higher RPM...(presumably as the centrifugal effect of the drilling has an impact ???)

Balance some discussion on CFD and it's limitations.

 

[MolaKule]

Interesting paper:

Quote:
P10/11, shows crank drilling changes on a 3 cylinder engine that allows changes in bearing design, offering a 15% reduction in crank oil supply requirement.

P12 to 14, shows oil distribution volumes in a 4 cylinder diesel. Then same crank drilling changes as previous section, and measured results...45% reduction in crank flows possible through drilling/bearing changes.


Yes, but it appears any reduction in oil flow also yields a temperature rise in the bearing and a rise in the oil temperature leaking from oil leakage past the bearings.

SO I guess the question is, was this simply an experiment to determine drilling positions verses temperature of exiting oil?

 

[jaj]

It reads more like an advertising piece for their new version of software. They've added some pretty sophisticated analytics that can start with a physical model of the bearing, the engine structure and the oil itself and solve for important parameters like film thickness and oil flow. Pretty clever.

If you ignore how challenging that modeling project actually was, the rest is pretty simple - slower flow means hotter engine parts and hotter oil, but hotter oil means thinner oil, and that means faster flow. You just iterate until it settles.

 

[MolaKule]

Quote:
You just iterate until it settles.


Quite advanced software indeed.

You infuse the software with some PDE's representing the mechanico-thermo-fluid physics of the problem and let the software converge on a solution.

Apparently what they did to improve simulation was to "couple" the SABRE software with a Computational Fluids Dynamics (CFD) simulation package, which seems to be the main thrust of their presentation.

Quote:
The ‘standard’ oil flow prediction tools based on the ‘Reynolds’ equation (SABRE-M and SABRE-EHL) give good
results which correlate well with measurement for many cases.
However, limitations arise due to ‘assumed’ oil feed pressure boundary conditions. Particularly cases involving:
- Aeration-cavitation (high engine speed).
- Low oil feed pressure.
- Intermittent feed (big-end fed from partial groove main without a cross drilling).
 A Computational Fluid Dynamics (CFD) tool has been developed.
- This can be combined with SABRE-EHL to provide more accurate boundary conditions and oil flow predictions.

 

[Shannow]

Agree with the iterative process...doing it the old way, we picked a temperature/viscosity, ran through the design charts, sommerfeld numbers etc., and got a heat balance...if it didn't match the assumption, you put the new number in and crank up the HP 15C again.

I agree that they are pushing their (great by the looks of it) software, but my main point in putting it in "articles" instead of "white papers" is that BITOG is awash with people who believe that oil carries heat out of the bearings (and bearing surfaces) rather then the shearing of the oil being the primary contributor of heat in bearings.

This, my warmup thread, and others, I'm trying to get the message out that:
* there is a temperature rise within bearing surfaces, due to viscous friction.
* bearings often operate at higher than coolant temperatures and bulk oil temperatures.
* some of the heat from the bearings is transferred to the coolant, rather than the other way around.
* more revs equals more heat.

I doubt that these things need demonstrating to either of you, but there's a large field of posters who have been swayed by oil 101, and some common myths.